CN114065801B - Soil monitoring method and system based on neural network model and readable storage medium - Google Patents
Soil monitoring method and system based on neural network model and readable storage medium Download PDFInfo
- Publication number
- CN114065801B CN114065801B CN202111199642.5A CN202111199642A CN114065801B CN 114065801 B CN114065801 B CN 114065801B CN 202111199642 A CN202111199642 A CN 202111199642A CN 114065801 B CN114065801 B CN 114065801B
- Authority
- CN
- China
- Prior art keywords
- soil
- livestock
- grass
- information
- neural network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A40/00—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production
- Y02A40/70—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production in livestock or poultry
Abstract
The invention discloses a soil monitoring method, a soil monitoring system and a readable storage medium based on a neural network model, wherein the method comprises the following steps: acquiring pressure distribution information of the soil in a target area based on a preset pressure sensor, and inputting the pressure distribution information into a trained path recognition neural network model to obtain a livestock advancing path; identifying grass data information on the livestock traveling path based on the collected livestock image information; screening all grass swarms on the soil surface layer in the livestock image information by utilizing big data analysis to obtain a coordinate set of a target grass swarms; outputting a monitoring area based on the coordinate set of the target grass group, and calling a preset sensor group in the monitoring area to identify soil information of a corresponding position. The method can automatically acquire the advancing path of the livestock on the grassland based on the set CNN neural network model, and further acquire grass swarm information on the path to identify a target grass swarm so as to accurately acquire the position of the soil to be monitored for monitoring.
Description
Technical Field
The invention relates to the technical field of soil monitoring, in particular to a soil monitoring method and system based on a neural network model and a readable storage medium.
Background
Animal husbandry often represents a subsidiary industry of crop production in the early stage of economic development, namely, the so-called "post-yard animal husbandry", and gradually develops into a relatively independent industry in certain departments along with the development of economy, such as pasture and grazing, and the pasture and grazing also need to be upgraded and modified synchronously along with the increase of the demand of meat and milk foods.
The definition of the pasture and the pasture area in the pasture also needs to be divided, wherein, in order to increase the yield of animal husbandry, the regular maintenance and detection can be carried out on the pasture and the grass in the pasture, correspondingly, the monitoring can also be carried out on the meadow soil in order to obtain better pasture.
Disclosure of Invention
In view of the above problems, an object of the present invention is to provide a soil monitoring method, system and readable storage medium based on a neural network model, which can automatically obtain the traveling path of livestock on the grassland based on a set CNN neural network model to identify a target grass group, and further accurately obtain the position of the soil to be monitored for monitoring.
The invention provides a soil monitoring method based on a neural network model in a first aspect, which comprises the following steps:
acquiring pressure distribution information of the soil in a target area based on a preset pressure sensor, and inputting the pressure distribution information into a trained path recognition neural network model to obtain a livestock advancing path;
identifying grass data information of the soil surface layer on the livestock traveling path based on the collected livestock image information;
screening all grass swarms on the soil surface layer in the livestock image information by utilizing big data analysis to obtain a coordinate set of a target grass swarms in the target area;
outputting a monitoring area based on the coordinate set of the target grass group, and calling a preset sensor group in the monitoring area to identify soil information of a corresponding position.
In this scheme, based on preset pressure sensor obtain in the target area the pressure distribution information of soil, input the route discernment neural network model that trains well, obtain the livestock route of marcing, specifically do:
establishing a communication connection with the pressure sensor based on a preset period;
acquiring a soil pressure value acquired by the pressure sensor to obtain pressure distribution information of the soil;
and inputting the pressure distribution information into the trained path recognition neural network model to obtain a simulation output value, and obtaining the livestock advancing path based on the simulation output value.
In this scheme, the path recognition neural network model training method is as follows:
acquiring a soil pressure value and soil pressure distribution information of historical detection data;
preprocessing the soil pressure value and the soil pressure distribution information of the historical detection data to obtain a training sample set;
inputting the training sample set into the initialized path recognition neural network model for training;
acquiring the accuracy of an output result;
and if the accuracy is greater than a preset accuracy threshold, stopping training to obtain the path recognition neural network model.
In this scheme, the livestock image information based on gathering discerns the grass crowd data message on the soil top layer on the livestock route of marcing is specifically:
establishing communication connection with a preset unmanned aerial vehicle, and acquiring the livestock image information based on the period;
acquiring position information corresponding to the livestock image information based on the livestock traveling path;
and acquiring corresponding grass group data information of the soil surface layer based on the position information.
In this scheme, the screening of all grass swarms on the soil surface layer in the livestock image information by utilizing big data analysis to obtain a coordinate set of a target grass swarms in the target area specifically includes:
after the grass data information on the livestock advancing path is obtained, screening all grass swarms in the livestock image information by utilizing big data analysis;
outputting the result after the screening is successful as the target grass group;
and extracting the position coordinates of each target grass group to obtain a coordinate set of the target grass group in the target area.
In this scheme, the monitoring area is output based on the coordinate set of the target grass group, and the soil information of the corresponding position is identified by calling the preset sensor group in the monitoring area, specifically:
outputting the monitoring area based on the target area after the coordinate set is obtained;
establishing communication connection with the sensor group in the monitoring area, and acquiring data acquired by each sensor in the sensor group based on the period, wherein the sensor group comprises but is not limited to a soil humidity sensor, a soil PH sensor and an inorganic element sensor;
and identifying the soil information of the corresponding position based on the data collected by each sensor.
The second aspect of the present invention further provides a soil monitoring system based on a neural network model, which includes a memory and a processor, wherein the memory includes a soil monitoring method program based on the neural network model, and when the soil monitoring method program based on the neural network model is executed by the processor, the following steps are implemented:
acquiring pressure distribution information of the soil in a target area based on a preset pressure sensor, and inputting the pressure distribution information into a trained path recognition neural network model to obtain a livestock advancing path;
identifying grass data information of the soil surface layer on the livestock traveling path based on the collected livestock image information;
screening all grass swarms on the soil surface layer in the livestock image information by utilizing big data analysis to obtain a coordinate set of a target grass swarms in the target area;
outputting a monitoring area based on the coordinate set of the target grass group, and calling a preset sensor group in the monitoring area to identify soil information of a corresponding position.
In this scheme, based on preset pressure sensor obtain in the target area the pressure distribution information of soil, input the route discernment neural network model that trains well, obtain the livestock route of marcing, specifically do:
establishing a communication connection with the pressure sensor based on a preset period;
acquiring a soil pressure value acquired by the pressure sensor to obtain pressure distribution information of the soil;
and inputting the pressure distribution information into the trained path recognition neural network model to obtain a simulation output value, and obtaining the livestock advancing path based on the simulation output value.
In this scheme, the path recognition neural network model training method is as follows:
acquiring a soil pressure value and soil pressure distribution information of historical detection data;
preprocessing the soil pressure value and the soil pressure distribution information of the historical detection data to obtain a training sample set;
inputting the training sample set into the initialized path recognition neural network model for training;
acquiring the accuracy of an output result;
and if the accuracy is greater than a preset accuracy threshold, stopping training to obtain the path recognition neural network model.
In this scheme, the livestock image information based on gathering discerns the grass crowd data message on the soil top layer on the livestock route of marcing is specifically:
establishing communication connection with a preset unmanned aerial vehicle, and acquiring the livestock image information based on the period;
acquiring position information corresponding to the livestock image information based on the livestock traveling path;
and acquiring corresponding grass group data information of the soil surface layer based on the position information.
In this scheme, the screening of all grass swarms on the soil surface layer in the livestock image information by utilizing big data analysis to obtain a coordinate set of a target grass swarms in the target area specifically includes:
after grass group data information on the livestock traveling path is obtained, screening all grass groups in the livestock image information by utilizing big data analysis;
outputting the result after the screening is successful as the target grass group;
and extracting the position coordinates of each target grass group to obtain a coordinate set of the target grass groups in the target area.
In this scheme, the monitoring area is output based on the coordinate set of the target grass group, and the soil information of the corresponding position is identified by calling the preset sensor group in the monitoring area, specifically:
outputting the monitoring area based on the target area after the coordinate set is obtained;
establishing communication connection with the sensor group in the monitoring area, and acquiring data acquired by each sensor in the sensor group based on the period, wherein the sensor group comprises but is not limited to a soil humidity sensor, a soil PH sensor and an inorganic element sensor;
and identifying the soil information of the corresponding position based on the data collected by each sensor.
A third aspect of the present invention provides a computer-readable storage medium, which includes a neural network model-based soil monitoring method program of a machine, and when the neural network model-based soil monitoring method program is executed by a processor, the method includes any one of the above steps.
The soil monitoring method, the soil monitoring system and the readable storage medium based on the neural network model can automatically acquire the traveling path of livestock on the grassland based on the set CNN neural network model, further acquire grass group information of eating preference of the livestock on the path to identify target grass groups in a livestock area, and can accurately acquire the position of soil to be monitored based on the position information of the target grass groups so as to monitor the soil by using a preset sensor.
Drawings
FIG. 1 is a flow chart of a soil monitoring method based on a neural network model according to the present invention;
fig. 2 shows a block diagram of a soil monitoring system based on a neural network model according to the present invention.
Detailed Description
In order that the above objects, features and advantages of the present invention can be more clearly understood, a more particular description of the invention will be rendered by reference to the appended drawings. It should be noted that the embodiments and features of the embodiments of the present application may be combined with each other without conflict.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, however, the present invention may be practiced in other ways than those specifically described herein, and therefore the scope of the present invention is not limited by the specific embodiments disclosed below.
Fig. 1 shows a flow chart of a soil monitoring method based on a neural network model according to the present application.
As shown in fig. 1, the present application discloses a soil monitoring method based on a neural network model, comprising the following steps:
s102, acquiring pressure distribution information of the soil in a target area based on a preset pressure sensor, and inputting the information into a trained path recognition neural network model to obtain a livestock advancing path;
s104, identifying grass data information of the soil surface layer on the livestock advancing path based on the collected livestock image information;
s106, screening all grass swarms on the soil surface layer in the livestock image information by utilizing big data analysis to obtain a coordinate set of a target grass swarms in the target area;
and S108, outputting a monitoring area based on the coordinate set of the target grass group, and calling a preset sensor group in the monitoring area to identify soil information at a corresponding position.
The method includes the steps of firstly obtaining pressure distribution information of soil in a target area through a set pressure sensor, inputting the pressure distribution information into a trained path recognition neural network model for training, obtaining a livestock advancing path based on a result output by the model, identifying grass data information on the advancing path according to a livestock image acquired by an unmanned aerial vehicle after obtaining the advancing path, obtaining feeding preference data information of livestock, further extracting grass data which are in accordance with livestock feeding in all grass swarms of a whole pasture through big data analysis to obtain a target grass swarms, further obtaining a coordinate set of the target grass swarms, outputting a monitoring area based on the coordinate set of the target grass swarms, and obtaining distribution data of preset sensors according to a distribution algorithm called by the monitoring area, wherein the preset sensors are soil sensors, and starting the soil sensors at corresponding positions to identify soil information at corresponding positions.
It is worth mentioning that the target area is a test area for experiments performed to obtain feeding preferences of the livestock, correspondingly, the pressure sensor distribution area is the test area and not the whole pasture, and the livestock image is image information of the whole pasture, including image information of the test area.
According to the embodiment of the invention, the pressure distribution information of the soil in the target area is obtained based on the preset pressure sensor and is input into the trained path recognition neural network model to obtain the advancing path of the livestock, and the method specifically comprises the following steps:
establishing a communication connection with the pressure sensor based on a preset period;
acquiring a soil pressure value acquired by the pressure sensor to obtain pressure distribution information of the soil;
and inputting the pressure distribution information into the trained path recognition neural network model to obtain a simulation output value, and obtaining the livestock advancing path based on the simulation output value.
It should be noted that the period may be set to "8" hours, pressure acquisition data of the pressure sensor within "8" hours is acquired to acquire a soil pressure value acquired by the pressure sensor, so as to obtain pressure distribution information of the soil, after the pressure distribution information is acquired, the pressure distribution information is input into the trained path recognition neural network model to obtain a simulation output value, and the livestock advancing path is obtained based on the simulation output value.
According to the embodiment of the invention, the path recognition neural network model training method comprises the following steps:
acquiring a soil pressure value and soil pressure distribution information of historical detection data;
preprocessing the soil pressure value and the soil pressure distribution information of the historical detection data to obtain a training sample set;
inputting the training sample set into the initialized path recognition neural network model for training;
acquiring the accuracy of an output result;
and if the accuracy is greater than a preset accuracy threshold, stopping training to obtain the path recognition neural network model.
It should be noted that path identification neural network model needs a large amount of historical data to train, and the data volume is big more, then the result is more accurate, in this application path identification neural network model can train as the input through the soil pressure value and the soil pressure distribution information of historical detection data, certainly, when carrying out neural network model training, not only will train through the soil pressure value and the soil pressure distribution information of historical detection data, still need combine definite livestock type to train, compare through a large amount of test data and true data, the result that obtains also can be more accurate, and then make the output result of abnormal cause neural network more accurate. Preferably, the accuracy threshold is generally set to 90%.
According to the embodiment of the invention, the identifying of the grass data information of the soil surface layer on the livestock advancing path based on the collected livestock image information specifically comprises:
establishing communication connection with a preset unmanned aerial vehicle, and acquiring the livestock image information based on the period;
acquiring position information corresponding to the livestock image information based on the livestock traveling path;
and acquiring corresponding grass group data information of the soil surface layer based on the position information.
It should be noted that the period is "8" hours, the livestock image information is an image within "8" hours, the position information of the traveling path on the livestock image information is obtained through the traveling path of the identified livestock, and grass flock data information of the corresponding position is obtained through the position information, so as to obtain the feed data information of the livestock.
According to the embodiment of the present invention, the screening of all grass swarms on the soil surface layer in the livestock image information by utilizing big data analysis to obtain the coordinate set of the target grass swarms in the target area specifically includes:
after the grass data information on the livestock advancing path is obtained, screening all grass swarms in the livestock image information by utilizing big data analysis;
outputting the result after the screening is successful as the target grass group;
and extracting the position coordinates of each target grass group to obtain a coordinate set of the target grass groups in the target area.
It should be noted that after the feeding information of the livestock is identified, the same grass group in all grass groups in the whole pasture is obtained by utilizing big data analysis to be used as the target grass group, and the position coordinates of each target grass group are extracted to obtain the coordinate set of the target grass group in the pasture.
According to the embodiment of the invention, the outputting of the monitoring area based on the coordinate set of the target grass group and the calling of the preset sensor group in the monitoring area to identify the soil information of the corresponding position specifically comprise:
outputting the monitoring area based on the target area after the coordinate set is obtained;
establishing communication connection with the sensor group in the monitoring area, and acquiring data acquired by each sensor in the sensor group based on the period, wherein the sensor group comprises but is not limited to a soil humidity sensor, a soil PH sensor and an inorganic element sensor;
and identifying the soil information of the corresponding position based on the data collected by each sensor.
It should be noted that the sensor group includes, but is not limited to, the soil moisture sensor, the soil PH sensor, the inorganic element sensor, and the soil moisture sensor, and the like, and the data collected by each sensor can identify the soil information at the corresponding position.
It is worth mentioning that the distribution algorithm is called to obtain the distribution data of the preset sensor, specifically:
acquiring a monitoring area based on the monitoring area;
obtaining an order of magnitude using the monitored area as an input to the distribution algorithm;
and acquiring distribution data of the preset sensor based on the magnitude.
It should be noted that the monitoring area is an irregular combination of scattered grass group and continuous grass group, so that in actual setting, the grass group area needs to be reasonably divided, and the calculation formula of the distribution algorithm is as follows:
wherein N is of said order of magnitude, S c Is the area of the continuous grass group, S d And for the area of the scattered grass group, alpha and beta are set grass group compensation factors, the order of magnitude N can be obtained through the distribution algorithm, and then the distribution data of the preset sensor is obtained according to the order of magnitude N.
It is worth mentioning that the method further comprises obtaining the physical condition information of the livestock according to the soil information, specifically:
identifying livestock manure position information based on the livestock image;
acquiring soil information data of corresponding positions based on the sensor group;
and identifying the physical condition of the livestock through the soil information data.
It should be noted that, in the process of raising livestock, it is necessary to treat and prevent animal diseases by using antibiotics, for example, fluoroquinolone (FQS) and Tetracycline (TCs) antibiotics are chemically synthesized antibiotics, most of the antibiotics cannot be completely absorbed by the body during the use, about 60% -90% of the antibiotics enter the environment in the form of their original forms or metabolites through feces and urine of livestock and poultry, and cause serious pollution to surface water, farmland soil and even underground water, so the current physical condition of the livestock can be identified by examining the feces data of livestock in the soil.
Fig. 2 shows a block diagram of a soil monitoring system based on a neural network model according to the present invention.
As shown in fig. 2, the invention discloses a soil monitoring system based on a neural network model, which includes a memory and a processor, wherein the memory includes a soil monitoring method program based on the neural network model, and the soil monitoring method program based on the neural network model implements the following steps when being executed by the processor:
acquiring pressure distribution information of the soil in a target area based on a preset pressure sensor, and inputting the pressure distribution information into a trained path recognition neural network model to obtain a livestock advancing path;
identifying grass data information of the soil surface layer on the livestock traveling path based on the collected livestock image information;
screening all grass swarms on the soil surface layer in the livestock image information by utilizing big data analysis to obtain a coordinate set of a target grass swarms in the target area;
outputting a monitoring area based on the coordinate set of the target grass group, and calling a preset sensor group in the monitoring area to identify soil information of a corresponding position.
The method includes the steps of firstly obtaining pressure distribution information of soil in a target area through a set pressure sensor, inputting the pressure distribution information into a trained path recognition neural network model for training, obtaining a livestock advancing path based on a result output by the model, identifying grass data information on the advancing path according to a livestock image acquired by an unmanned aerial vehicle after obtaining the advancing path, obtaining feeding preference data information of livestock, further extracting grass data which are in accordance with livestock feeding in all grass swarms of a whole pasture through big data analysis to obtain a target grass swarms, further obtaining a coordinate set of the target grass swarms, outputting a monitoring area based on the coordinate set of the target grass swarms, and obtaining distribution data of preset sensors according to a distribution algorithm called by the monitoring area, wherein the preset sensors are soil sensors, and starting the soil sensors at corresponding positions to identify soil information at corresponding positions.
It is worth mentioning that the target area is an experimental area for experiments performed to obtain feeding preferences of livestock, accordingly, the pressure sensor distribution area is the experimental area rather than a whole pasture, and the livestock image is image information of the whole pasture, including image information of the experimental area.
According to the embodiment of the invention, the pressure distribution information of the soil in the target area is obtained based on the preset pressure sensor and is input into the trained path recognition neural network model to obtain the advancing path of the livestock, and the method specifically comprises the following steps:
establishing a communication connection with the pressure sensor based on a preset period;
acquiring a soil pressure value acquired by the pressure sensor to obtain pressure distribution information of the soil;
and inputting the pressure distribution information into the trained path recognition neural network model to obtain a simulation output value, and obtaining the livestock advancing path based on the simulation output value.
It should be noted that the period may be set to "8" hours, pressure acquisition data of the pressure sensor within "8" hours is acquired to acquire a soil pressure value acquired by the pressure sensor, so as to obtain pressure distribution information of the soil, after the pressure distribution information is acquired, the pressure distribution information is input into the trained path recognition neural network model to obtain a simulation output value, and the livestock advancing path is obtained based on the simulation output value.
According to the embodiment of the invention, the path recognition neural network model training method comprises the following steps:
acquiring a soil pressure value and soil pressure distribution information of historical detection data;
preprocessing the soil pressure value and the soil pressure distribution information of the historical detection data to obtain a training sample set;
inputting the training sample set into the initialized path recognition neural network model for training;
acquiring the accuracy of an output result;
and if the accuracy is greater than a preset accuracy threshold, stopping training to obtain the path recognition neural network model.
It should be noted that path identification neural network model needs a large amount of historical data to train, and the data volume is big more, and then the result is more accurate, in this application path identification neural network model can train as the input through the soil pressure value and the soil pressure distribution information of historical detection data, certainly, when carrying out neural network model training, not only will train through the soil pressure value and the soil pressure distribution information of historical detection data, still need combine definite livestock type to train, compare through a large amount of test data and true data, the result that obtains also can be more accurate, and then makes the output result of unusual reason neural network more accurate. Preferably, the accuracy threshold is generally set to 90%.
According to the embodiment of the invention, the identifying of the grass data information of the soil surface layer on the livestock advancing path based on the collected livestock image information specifically comprises:
establishing communication connection with a preset unmanned aerial vehicle, and acquiring the livestock image information based on the period;
acquiring position information corresponding to the livestock image information based on the livestock traveling path;
and acquiring corresponding grass group data information of the soil surface layer based on the position information.
It should be noted that the period is "8" hours, the livestock image information is an image within "8" hours, the position information of the traveling path on the livestock image information is obtained through the traveling path of the identified livestock, and grass flock data information of the corresponding position is obtained through the position information, so as to obtain the feed data information of the livestock.
According to the embodiment of the present invention, the screening of all grass swarms on the soil surface layer in the livestock image information by utilizing big data analysis to obtain the coordinate set of the target grass swarms in the target area specifically includes:
after the grass data information on the livestock advancing path is obtained, screening all grass swarms in the livestock image information by utilizing big data analysis;
outputting the result after the screening is successful as the target grass group;
and extracting the position coordinates of each target grass group to obtain a coordinate set of the target grass groups in the target area.
It should be noted that after the feeding information of the livestock is identified, the same grass group in all grass groups in the whole pasture is obtained by utilizing big data analysis to be used as the target grass group, and the position coordinates of each target grass group are extracted to obtain the coordinate set of the target grass group in the pasture.
According to the embodiment of the invention, the outputting of the monitoring area based on the coordinate set of the target grass group and the calling of the preset sensor group in the monitoring area to identify the soil information of the corresponding position specifically comprise:
outputting the monitoring area based on the target area after the coordinate set is obtained;
establishing communication connection with the sensor group in the monitoring area, and acquiring data acquired by each sensor in the sensor group based on the period, wherein the sensor group comprises but is not limited to a soil humidity sensor, a soil PH sensor and an inorganic element sensor;
and identifying the soil information of the corresponding position based on the data collected by each sensor.
It should be noted that the sensor group includes, but is not limited to, the soil moisture sensor, the soil PH sensor, the inorganic element sensor, and the like, and the data collected by each sensor may identify the soil information at the corresponding position.
It is worth mentioning that the distribution algorithm is called to obtain the distribution data of the preset sensor, specifically:
acquiring a monitoring area based on the monitoring area;
obtaining an order of magnitude using the monitored area as an input to the distribution algorithm;
and acquiring distribution data of the preset sensors based on the magnitude.
It should be noted that the monitoring area is an irregular combination of scattered grass group and continuous grass group, so that in actual setting, the grass group area needs to be reasonably divided, and the calculation formula of the distribution algorithm is as follows:
wherein N is of said order of magnitude, S c Is the area of the continuous grass group, S d And for the area of the scattered grass group, alpha and beta are set grass group compensation factors, the order of magnitude N can be obtained through the distribution algorithm, and then the distribution data of the preset sensor is obtained according to the order of magnitude N.
It is worth mentioning that the method further comprises obtaining the physical condition information of the livestock according to the soil information, specifically:
identifying livestock manure position information based on the livestock image;
acquiring soil information data of corresponding positions based on the sensor group;
and identifying the physical condition of the livestock through the soil information data.
It should be noted that, in the process of raising livestock, it is necessary to treat and prevent animal diseases by using antibiotics, for example, fluoroquinolone (FQS) and Tetracycline (TCs) antibiotics are chemically synthesized antibiotics, most of the antibiotics cannot be completely absorbed by the body during the use, about 60% -90% of the antibiotics enter the environment in the form of their original forms or metabolites through feces and urine of livestock and poultry, and cause serious pollution to surface water, farmland soil and even underground water, so the current physical condition of the livestock can be identified by examining the feces data of livestock in the soil.
A third aspect of the present invention provides a computer-readable storage medium, which includes a neural network model-based soil monitoring method program of a machine, and when the neural network model-based soil monitoring method program is executed by a processor, the method includes any one of the above steps.
The soil monitoring method, the soil monitoring system and the readable storage medium based on the neural network model can automatically acquire the traveling path of livestock on the grassland based on the set CNN neural network model, further acquire grass group information of eating preference of the livestock on the path to identify target grass groups in a livestock area, and can accurately acquire the position of soil to be monitored based on the position information of the target grass groups so as to monitor the soil by using a preset sensor.
In the several embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The above-described device embodiments are merely illustrative, for example, the division of the unit is only a logical functional division, and there may be other division ways in actual implementation, such as: multiple units or components may be combined, or may be integrated into another system, or some features may be omitted, or not implemented. In addition, the coupling, direct coupling or communication connection between the components shown or discussed may be through some interfaces, and the indirect coupling or communication connection between the devices or units may be electrical, mechanical or other forms.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units; can be located in one place or distributed on a plurality of network units; some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, all the functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately regarded as one unit, or two or more units may be integrated into one unit; the integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional unit.
Those of ordinary skill in the art will understand that: all or part of the steps of implementing the method embodiments may be implemented by hardware related to program instructions, and the program may be stored in a computer-readable storage medium, and when executed, executes the steps including the method embodiments; and the aforementioned storage medium includes: a mobile storage device, a Read-only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
Alternatively, the integrated unit of the present invention may be stored in a computer-readable storage medium if it is implemented in the form of a software functional module and sold or used as a separate product. Based on such understanding, the technical solutions of the embodiments of the present invention may be essentially implemented or a part contributing to the prior art may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the methods described in the embodiments of the present invention. And the aforementioned storage medium includes: a removable storage device, a ROM, a RAM, a magnetic or optical disk, or various other media capable of storing program code.
Claims (6)
1. A soil monitoring method based on a neural network model is characterized by comprising the following steps: acquiring pressure distribution information of the soil in a target area based on a preset pressure sensor, and inputting the pressure distribution information into a trained path recognition neural network model to obtain a livestock advancing path;
identifying grass data information of the soil surface layer on the livestock traveling path based on the collected livestock image information;
screening all grass swarms on the soil surface layer in the livestock image information by utilizing big data analysis to obtain a coordinate set of a target grass swarms in the target area;
outputting a monitoring area based on the coordinate set of the target grass group, and calling a preset sensor group in the monitoring area to identify soil information at a corresponding position;
the method comprises the following steps of acquiring pressure distribution information of soil in a target area based on a preset pressure sensor, inputting the pressure distribution information into a trained path recognition neural network model to obtain a livestock advancing path, and specifically comprising the following steps: establishing communication connection with the pressure sensor based on a preset period;
acquiring a soil pressure value acquired by the pressure sensor to obtain pressure distribution information of the soil;
inputting the pressure distribution information into the trained path recognition neural network model to obtain a simulation output value, and obtaining the livestock advancing path based on the simulation output value;
the grass data information of the soil surface layer on the livestock advancing path is identified based on the collected livestock image information, and the method specifically comprises the following steps: establishing communication connection with a preset unmanned aerial vehicle, and acquiring the livestock image information based on the period;
acquiring position information corresponding to the livestock image information based on the livestock traveling path;
acquiring grass group data information of the corresponding soil surface layer based on the position information;
the big data analysis is used for screening all grass swarms on the soil surface layer in the livestock image information to obtain a coordinate set of target grass swarms in the target area, and the method specifically comprises the following steps: after the grass data information on the livestock advancing path is obtained, screening all grass swarms in the livestock image information by utilizing big data analysis;
outputting the result after the screening is successful as the target grass group;
and extracting the position coordinates of each target grass group to obtain a coordinate set of the target grass groups in the target area.
2. The soil monitoring method based on the neural network model as claimed in claim 1, wherein the path recognition neural network model training method is: acquiring a soil pressure value and soil pressure distribution information of historical detection data;
preprocessing the soil pressure value and the soil pressure distribution information of the historical detection data to obtain a training sample set;
inputting the training sample set into the initialized path recognition neural network model for training;
acquiring the accuracy of an output result;
and if the accuracy is greater than a preset accuracy threshold, stopping training to obtain the path recognition neural network model.
3. The soil monitoring method based on the neural network model as claimed in claim 1, wherein the monitoring area is output based on the coordinate set of the target grass group, and a preset sensor group in the monitoring area is called to identify soil information of a corresponding position, specifically: outputting the monitoring area based on the target area after the coordinate set is obtained;
establishing communication connection with the sensor group in the monitoring area, and acquiring data acquired by each sensor in the sensor group based on the period, wherein the sensor group comprises but is not limited to a soil humidity sensor, a soil PH sensor and an inorganic element sensor;
and identifying the soil information of the corresponding position based on the data collected by each sensor.
4. A soil monitoring system based on a neural network model is characterized by comprising a memory and a processor, wherein the memory comprises a soil monitoring method program based on the neural network model, and the soil monitoring method program based on the neural network model realizes the following steps when being executed by the processor: acquiring pressure distribution information of the soil in a target area based on a preset pressure sensor, and inputting the pressure distribution information into a trained path recognition neural network model to obtain a livestock advancing path;
identifying grass data information of the soil surface layer on the livestock traveling path based on the collected livestock image information;
screening all grass swarms on the soil surface layer in the livestock image information by utilizing big data analysis to obtain a coordinate set of a target grass swarms in the target area;
outputting a monitoring area based on the coordinate set of the target grass group, and calling a preset sensor group in the monitoring area to identify soil information at a corresponding position;
the method comprises the following steps of acquiring pressure distribution information of soil in a target area based on a preset pressure sensor, inputting the pressure distribution information into a trained path recognition neural network model to obtain a livestock advancing path, and specifically comprising the following steps: establishing a communication connection with the pressure sensor based on a preset period;
acquiring a soil pressure value acquired by the pressure sensor to obtain pressure distribution information of the soil;
inputting the pressure distribution information into the trained path recognition neural network model to obtain a simulation output value, and obtaining the livestock advancing path based on the simulation output value;
the livestock image information based on the collection is used for identifying the grass data information of the soil surface layer on the livestock advancing path, and the method specifically comprises the following steps: establishing communication connection with a preset unmanned aerial vehicle, and acquiring the livestock image information based on the period;
acquiring position information corresponding to the livestock image information based on the livestock traveling path;
acquiring grass group data information of the corresponding soil surface layer based on the position information;
the big data analysis is used for screening all grass swarms on the soil surface layer in the livestock image information to obtain a coordinate set of target grass swarms in the target area, and the method specifically comprises the following steps: after the grass data information on the livestock advancing path is obtained, screening all grass swarms in the livestock image information by utilizing big data analysis;
outputting the result after the screening is successful as the target grass group;
and extracting the position coordinates of each target grass group to obtain a coordinate set of the target grass groups in the target area.
5. The soil monitoring system based on the neural network model as claimed in claim 4, wherein the path recognition neural network model training method is: acquiring a soil pressure value and soil pressure distribution information of historical detection data;
preprocessing the soil pressure value and the soil pressure distribution information of the historical detection data to obtain a training sample set;
inputting the training sample set into the initialized path recognition neural network model for training;
acquiring the accuracy of an output result;
and if the accuracy is greater than a preset accuracy threshold, stopping training to obtain the path recognition neural network model.
6. A computer-readable storage medium, characterized in that the computer-readable storage medium includes a neural network model-based soil monitoring method program, and when the neural network model-based soil monitoring method program is executed by a processor, the steps of a neural network model-based soil monitoring method according to any one of claims 1 to 3 are implemented.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111199642.5A CN114065801B (en) | 2021-10-14 | 2021-10-14 | Soil monitoring method and system based on neural network model and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111199642.5A CN114065801B (en) | 2021-10-14 | 2021-10-14 | Soil monitoring method and system based on neural network model and readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114065801A CN114065801A (en) | 2022-02-18 |
| CN114065801B true CN114065801B (en) | 2022-10-28 |
Family
ID=80234516
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111199642.5A Active CN114065801B (en) | 2021-10-14 | 2021-10-14 | Soil monitoring method and system based on neural network model and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114065801B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116930459B (en) * | 2023-07-25 | 2024-03-12 | 江苏龙环环境科技有限公司 | Soil in-situ detection device and detection method thereof |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102034337A (en) * | 2009-09-25 | 2011-04-27 | 中国农业科学院农业资源与农业区划研究所 | System and method for prairie snow disaster remote sensing monitoring and disaster situation evaluation |
| CN110376138A (en) * | 2019-08-05 | 2019-10-25 | 北京绿土科技有限公司 | Land quality monitoring method based on Airborne Hyperspectral |
| CN110799030A (en) * | 2017-05-30 | 2020-02-14 | 沃尔塔机器人有限责任公司 | Method for controlling soil working device based on image processing and related system |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10772295B2 (en) * | 2016-08-22 | 2020-09-15 | International Business Machines Corporation | Unmanned aerial vehicle for determining geolocation foraging zones |
| CN109997116A (en) * | 2016-09-09 | 2019-07-09 | 沃尔玛阿波罗有限责任公司 | Device and method for monitoring scene |
| CN107220967A (en) * | 2017-05-08 | 2017-09-29 | 新疆农业大学 | A kind of grassland soil degradation evaluation method |
| CN108207653A (en) * | 2018-01-31 | 2018-06-29 | 佛山市神风航空科技有限公司 | A kind of grazing system based on unmanned plane |
| CN108535743A (en) * | 2018-04-16 | 2018-09-14 | 北京农业信息技术研究中心 | One kind putting herding positioning system and method in a suitable place to breed |
| CN110221588A (en) * | 2019-07-09 | 2019-09-10 | 哈尔滨商业大学 | A kind of wisdom cattle-breeding system based on Internet of Things |
| CN114070862B (en) * | 2021-10-15 | 2022-09-16 | 中国科学院地理科学与资源研究所 | Livestock raising control method and system based on Internet of things and readable storage medium |
-
2021
- 2021-10-14 CN CN202111199642.5A patent/CN114065801B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102034337A (en) * | 2009-09-25 | 2011-04-27 | 中国农业科学院农业资源与农业区划研究所 | System and method for prairie snow disaster remote sensing monitoring and disaster situation evaluation |
| CN110799030A (en) * | 2017-05-30 | 2020-02-14 | 沃尔塔机器人有限责任公司 | Method for controlling soil working device based on image processing and related system |
| CN110376138A (en) * | 2019-08-05 | 2019-10-25 | 北京绿土科技有限公司 | Land quality monitoring method based on Airborne Hyperspectral |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114065801A (en) | 2022-02-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Neethirajan | The role of sensors, big data and machine learning in modern animal farming | |
| Van Hertem et al. | Lameness detection based on multivariate continuous sensing of milk yield, rumination, and neck activity | |
| Jingqiu et al. | Cow behavior recognition based on image analysis and activities | |
| Maselyne et al. | Methods to construct feeding visits from RFID registrations of growing-finishing pigs at the feed trough | |
| Vázquez Diosdado et al. | Space-use patterns highlight behavioural differences linked to lameness, parity, and days in milk in barn-housed dairy cows | |
| Maselyne et al. | Online warning systems for individual fattening pigs based on their feeding pattern | |
| Tullo et al. | Precision livestock farming: an overview of image and sound labelling | |
| Van De Gucht et al. | Farm-specific economic value of automatic lameness detection systems in dairy cattle: From concepts to operational simulations | |
| Van Erp-van der Kooij et al. | Using precision farming to improve animal welfare. | |
| Tekin et al. | Precision livestock farming technologies: Novel direction of information flow | |
| US20210076644A1 (en) | Animal health, behavior, and inventory monitoring | |
| Stygar et al. | Analyses of body weight patterns in growing pigs: a new view on body weight in pigs for frequent monitoring | |
| CN114065801B (en) | Soil monitoring method and system based on neural network model and readable storage medium | |
| Visser et al. | Phenomics for sustainable production in the South African dairy and beef cattle industry | |
| Dittrich et al. | Variable selection for monitoring sickness behavior in lactating dairy cattle with the application of control charts | |
| Liu et al. | Classification of cow behavior patterns using inertial measurement units and a fully convolutional network model | |
| Tzanidakis et al. | Precision Livestock Farming (PLF) systems: Improving sustainability and efficiency of animal production | |
| CN113516139A (en) | Data processing method, device, equipment and storage medium | |
| Ojukwu et al. | Development of a computer vision system to detect inactivity in group-housed pigs | |
| CN114070862B (en) | Livestock raising control method and system based on Internet of things and readable storage medium | |
| Cappai et al. | Integrating the RFID identification system for Charolaise breeding bulls with 3D imaging for virtual archive creation | |
| AU2020101719A4 (en) | An artificial intelligence and internet of things based automated system for animal health care | |
| Neethirajan | Digitalization of animal farming | |
| Malhotra et al. | Application of AI/ML approaches for livestock improvement and management | |
| CN113971227B (en) | Big data based livestock monitoring method, system and readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |